The Isotope Effect in the Fixation of Nitrogen by Azotobacter

(1) (a) Geophysical Laboratory, Carnegie Institution of Washing- ton, Washington, D C ; (b) Continental Oil Company, Ponca City,. Oklahoma. (2) T C Ho...
0 downloads 0 Views 360KB Size
c.

376

THOMASHOERING AND [CONTRIBUTION FROM

THE

HARRELL

T. FORD

DEPARTMENT O F CHEMISTRY,

Vol. 82

UNIVERSITY OB

ARKAXSAS]

The Isotope Effect in the Fixation of Nitrogen by Azotobacter E3Y

TrIohfAS e. HOE RING^"

AND

HARRBIL T. FORD'^

RECEIVEDAPRIL8, 1969 The relative rates of fixation of N14N14and W N 1 4 by Azotobacter have been studied. The ratio of these rates is 1.000 & 0.001. The isotope effect in a heterogeneous reaction, the catalytic hydrogenation of the double bond in azobenzene, has been studied also. N o isotope effect was found. The rate-determining step in the mechanism of these reactions does not involve a change in bonding t o nitrogen. The geochemical implications are discussed.

Introduction The biological fixation of molecular nitrogen is a very important process from many points of view and information on the mechanism of this reaction is of interest. This paper reports on the relative rates of biological fixation of the isotopic forms of molecular nitrogen. Isotope effects in biological systems have been reported for several elements including hydrogen, carbon, oxygen and sulfur. These isotope effects are important in determining the distribution of the stable isotopes of these elements in nature. Experiments on the isotope effect in the heterogeneous hydrogenation of azobenzene to hydrazobenzene were undertaken to see if there were any similarities between the mechanism of this reaction and the fixation reaction.

Experimental Burk's nitrogen-free media was prepared by adding the following to each liter of distilled water: 0.2 g. of K&POr, 0.8 g. of K2HP04, 0.2 g. of hlgSOa.iH20, 0.1 g. of CaS04. 2H20, 20 g. of sucrose, 1 mg. of ferric sulfate and 0.1 g. of molybdic acid. Two hundred milliliter volumes of the medium was p l x e d in 500-ml. Kjeldahl flasks and sterilized a t 18 pounds pressure for 20 to 30 minutes. The original stocks of Azotobacter agile, Azotobacter chroococcum, Azotobacter indzcum and Azotobacter vznelandii were obtained from American Type Culture Collection, Washington, D . C. The flasks were innoculated by sterile techniques and stored for growth and fixation at room temperature out of the direct sunlight. Growth was quenched at varying intervals by adding concentrated sulfuric acid, and nitrogen determinations were made by the Kjeldahl method. Innoculated blank samples showed less than 0.3 mg. nitrogen present. The ammonium sulfate isolated from the Kjeldahl analysis was used for the isotopic analysis of the fixed nitrogen. The procedures for the isotopic analysis of nitrogen have been described previously.2 Since the Azotobacter used only an infinitesimal fraction of the N2 available to them, the kinetic isotope effect p, which is defined as 8 = 2 (1) r15

is given by

-

p = - K. Rr where r14 and are the rates of fixation of W 4 P 4and SW14, respectively, and R, and Rf are the ratios of hT15 t o NI4in the air and in the fixed nitrogen, respectively. Pure trans-azobenzene was prepared by the elution of a solution of commercially available azobenzene from a chromatography column of alumina with petroleum ether.3 h platinum-charcoal catalyst was prepared by the method described by B a l t ~ l y . ~It has been shown4that azobenzene in sodium methoxide reacts with hydrogen in the presence of platinized charcoal t o yield only hydrazobenzene. (1) (a) Geophysical Laboratory, Carnegie Institution of Washington, Washington, D C ; (b) Continental Oil Company, Ponca City, Oklahoma. (2) T C Hoering, Science, 122, 1233 (1955). ( . I ) A H Cook, J Chem SOL.,833 (10381 (41 R Baltzly, T m s J O C R N A ~ . 74, 4.556 (l9W).

In these experiments, samples of azobenzene were partially hydrogenated under the conditions mentioned. The reaction mixture was then acidified to permit the rearrangement of the hydrazobenzene to isomeric benzidines. The unreacted azobenzene was isolated by the chromatography method described previously. The samples of the starting azobenzene and the unreacted fraction from the hydrogendtion were treated with sodium hyposulfite and then subjected to a Kjeldahl digestion for the isolation of the nitrogen for isotopic analysis and for a measurement of the fraction reacted in the reduction reaction. The ammonium sulfate from the Kjeldahl analysis was converted t o N2 for isotopic analysis as before. Under these conditions, the kinetic isotope effect B is given by6

where f is the fraction of the azobenzene reacted, R., is the ratio of W5 t o XI4 in the starting azobenzene and R6. is the isotope ratio in the remaining azobenzene. All steps in this procedure were tested t o ensure quantitative conversion t o prevent isotope fractionation in the preparation of nitrogen for isotopic analysis. The results of these experiments are given in Tables I and 11. Within the precision of the experiments, both reactions have no isotope effect.

Discussion Since the net result of both chemical reactions studied is a change in the chemical bonding to nitrogen, an isotope effect may be expected. The kinetics of nitrogen fixation by Azotobacter have been studied by Wilson and co-workers6 and the data has been fitted to the rate law corresponding to the Michaelis-Menten expression E

+

EN,

ES2

K2

K1 rapid and

(4)

reversible

ka

E

+ Products

slow (5)

The isotope effects in a system with a weak preassociation have been discussed by Bigeleisen and Wolfsberg.' The observed isotope effect is the product of the isotope effects in step 4 multiplied by the effect in (5)

where the single and double primes refer t o NI4 and NI6 containing molecules, respectively. Lineweaver has shownRthat the interaction in step 4 is probably a weak one. Little isotope effect is ( 5 ) J. Y . Tong and P. Yankwich, J . P h y s . Chcm., 61, 540 (1957). (0) P. W. Wilson, R. H. Burris and C. J. Lind, Proc. Nall. Acnd. Sci , 28, 243 (1942). (7) J. Bigeleisen, J . C h o n . Phys., 17, 675 (1949); J. Bigeleisen and M. Wolfsberg in "Advances in Chemical Physics," Vol. I , edited by I. Prigogine, Chapter 11, "Theoretical and Experimental Aspects of Isotope Effects in Chemical Kinetics," Interscience Publishers, Inc., X e w York, S . Y . , 1958. 18) H. Lineweaver, .I. Bioi. Chrin , laZ, 549 (1938).

expected in such cases. For example, there is no isotope effect in the solubility of Ns in water.9 The nature of the chemical change in step 5 is not known. Evidence indicates that the first recognizable compound formed is ammonia, but the intermediates, X,Y,Z are not known. EN2 --f X +Y --f 2 --f SI33 E S2

TABLE I

THEISOTOPE EFFECT IN

THE FIXATION O F ATMOSPHERIC BY Azotobacter Nitrogen Fixation : fixed, mg. period, days1 B

(7)

Gestlo has summarized some very speculative possibilities for the reaction in (5). They are

‘i202

+

+Nz+

+e

ionization (8) N2 --+-2N dissociation (9) H f ?.T2 -+ HN=SH hydrogenation (IO) H10 Nz +HOS=NOH oxidation and hydrolysis (11)

+ +

Each of these reactions involves some change in the bonding to a nitrogen atom and thus may be accompanied by an isotope eKect. An attempt was made to estimate the magnitude of this isotope effect in reactions S to 10 by the method of Bigel e i ~ e n . ~The necessary data for vibrational frequencies of the nitrogen fourteen containing molecules has been tabulated by Herzberg. l1 The calculation of the isotope effect for reaction 10 was made using the nitrogen t o nitrogen double bond frequency of diazomethane and the nitrogen to hydrogen bond frequency of ammonia. The frequency shifts on isotopic substitution were calculated using the harmonic oscillator approximation and it was assumed that only the frequencies of the bonds directly connected with the isotopic center were affected by the substitution. No attempt was made to calculate the “effective mass” term of the Bigeleisen equation as this would require a greater knowledge of the geometry of the activated complex than is available. l 2 Only a value for the “temperature dependent” term of the equation was calculated. Since the “effective mass” term varies from one to values larger than one, the calculated isotope effect is a mininium value which is all that is required for the purposes of this study. The results are shown in Table 111. All predict a measurable isotope effect. The calculations for reaction 11 were not made. Since there is probably no isotope effect in step 4 and since any change in bonding to nitrogen in step 5 would probably have an isotope efiect, it is concluded that either (4) and (5) do not adequately describe the system or that the change in (5) is of an unknown kind that does not involve any change in bonding t o a nitrogen. The rate-determining step of the fixation reaction may be connected with the formation of the enzyme-substrate complex in a heterogeneous, catalytic process. m‘angI3 has shown that the isotope effect in the catalytic decomposition of (9) T. C. Hoeriag and H. E. Moore, Geochim. el Cosmochim. Acta, 13, 225 (1958). (10) H. Gest, J. Judis a n d H. D. Peck, J r . , “Inorganic Nitrogen Metabolism,” edited by W. D. McElroy and B. Glass, Johns Hopkins Press, Baltimore, hld., 1956. (11) G. Iierzberg, “Spectra of Diatomic Molecules,” D. Van Nostrand Co., New York, N. y.,1950; ‘‘Infra-Red and Raman Spectra,” D. Van Nostrand Co., New York, N.Y . , 1945. (12) J. Bigeleisen and h f . Wolfsberg, J . Chem. Phys., 2 1 , 1972 (lU.33); 2 2 . 12(j4 (1$l:~4). (1.3) J. H Wang, J Phys C h e m . , 69, 1115 (1955).

7.3 8.9 10.1 9.9 9.3 9.6

12 13 14 15 17 19

Azotobacter agile

+

iY2

377

ISOTOPE EFFECTIN FIXATION OF NITROGEN BY AZOTOBACTER

Jan. 20, 1960

0.9997 1.0014 1.0002 1.0013 1.0031 1.0035 .

Av.

15 17 19 21 25

1.5 1.9 3.2 3.2 4.6 14.6 12.8 14.6

38 38 38 4s 48 48 48

4.8 3.2 3.1 2.4 2.1 2.4 4.9

11 12

Azotobacter chroococcum

14

,9961

,9973 .9992 .9989 .9993

4.9 9.4 6.0 9.1 8.1 9.0 14.3

9 11 13 14 16 18 19

1.0007 0.9920 ,9910

Av.

Azotobacter uinelandii

1.0015 0.9992 0,9985 1.0009 1I0002 1.0004 1.0004 1.0019 1 0015

Av.

Azotobacter indicum

Nz

0.9963 1.0015 1 0012 1.0008 1.0030 1,0041 1.0031 1.0021

Av.

1.0022

TABLE I1 THE I s o r o m EFFECTI N THE CATALVTIC REDUCTION OF AZOBENZENE Fraction reacted

B

0 372 ,626

0.9965 ,9972 ,9971 1.0033 1.0032 1.0000 0.9973 1.0000 1.0000

c--

, 1 1 1

432 ,725 .945 ,637 ,809 ,857 Xv.

0.9994

TABLE 111 ISOTOPEEFFECTS CALCULATED FOR REACTIONS 8, 9 AT 25’

AXD

Reaction

Ratio of specific rate constants for N’d and Nlb compounds

(8) Ionization (9) Dissociation (10) Hydrogenation

1.006 1.084 1.027

10

hydrogen peroxide can varv with the size of the catalyst particle, being small for the diffusion of a ~~

THOMAS C. HOERING AND HARRELL T. FORD

378

reactant molecule up to a large, planar catalytic surface and being large for reactions with catalysts of molecular dimensions (a homogeneous reaction). An extension of the theory of Michaelis and Menten has been given by Chancel4 for the case where the reaction of the enzyme-substrate complex is second order EN2

E + ri~ EN? + AH2 +E + Products

(12)

(13)

Vol. 82

than the atmosphere. It is postulated that the kinetic processes of fixation and dc-nitrification for the isotopic forms of nitrogen can be schematically represented as in equation 14 biosphere

77----f

k:’

k1’ Atmosphere

kn’;

atmosphere (14)

ti

sediments If it is assumed that the formation of the reactant fixation de-nitrification AH2 is the slow step in the sequence rather than where the single and double primes indicate reaction 13, then the over-all process of nitrogen fixation by Azotobacter will not have an observ- specific rate constants for NI4 and ? J 1 5 containing able nitrogen isotope effect. This assumption molecules, respectively. Hutchinson’s inventoryl6 requires that the activation energy for other re- of terrestrial nitrogen, as modified by Scalan,I9 actants in the system is larger than for the breaking is shown in Table IV. I n equation 14, k2 is probof a very stable nitrogen to nitrogen bond in Nz. TABLE IT The anomalies between the large bond energy and INVENTORY O F TERRESTRIAL A-ITROGEX the rate of reaction of N2 has been pointed out by Grams nitrogen per cm.2 of Kamen.I5 Region t h e earth’s surface The sulfur isotope effect in the reduction of Core 26 sulfate ion by Desulphovibrio desulphuricans has Mantle 5660 been studied by Harrison and Thode.I6 They Crust 102 explained the occurrence or non-occurrence of an Atmosphere 755 isotope effect according to where the rate-deterSediments 67-108 mining step of their reaction sequence appeared. Nitrate deposits 2 x 10-6 The lack of an isotope effect in the catalytic Biosphere Very small hydrogenation of azobenzene can be explained similarly. The slow step in the process may be ably greater than kl; fixation is the slow step in connected with the adsorption of hydrogen. The the general process. A straightforward applicastep in which there is a change in bonding in the tion of the kinetic laws for successive, pseudonitrogens follows and is more rapid than the first-order reactions with the facts that kl’lkl” = 1 adsorption or formation of the catalyst-substrate and that the amount of N in the biosphere and complex. The carbon isotope effect in the catalytic sediments is less than the amount in the atmosphere hydrogenation of stilbene to bibenzyl was studied leads t o by Bonner and C01lins.l~ They found that the ratio of the rates of reduction of the CIZcompound to the C14 compound was 1.02 using Raney nickel where B’ and B” are the NI5and the N14content of in benzene or platinum in ethanol as the catalyst. the biosphere and sediments, A ’ and A ” are the N15 They also studied the relative rates of hydrogena- and N14 content of the atmosphere. Thus the tion of CI2 and CI4 acetophenone to l-phenyl- average Nl5-NI4 ratio in the biosphere and sediethanol. The ratio of the rates was 1.115 using ments is determined by the isotope effect in the platinum in ethanol as the catalyst and 1.0susing denitrification process and since lithium aluminum hydride in ether. I n these processes, the rate-determining step is the change of bonding a t the carbon undergoing isotopic substitution. then Biological nitrogen fixation is a very important geochemical process and represents the chief means by which atmospheric nitrogen enters the biosphere. If one assumes that symbiotic nitro- Isotope effects are to be expected in the circulation gen fixation has the same isotope effect as non- of nitrogen in the biosphere and differences in symbiotic, then this step partially sets the average nitrogen isotope ratios in biogenic materials have Nl5-NI4 ratio in the biosphere. Unpublished been observed. The isotope ratio calculated in results from this Laboratory show that the nitro- equation 15 is the average isotope content of the gen in naturally occurring amino acids and in biosphere and sediments. sedimentary rocks has a greater Nl5-Nl4 ratio Acknowledgments.-This work was supported by the U. S. Atomic Energy Commission through (14) B. Chance in “Techniques of Organic Chemistry,” Vol. VII. “Investigation of Rates and Mechanisms of Reactions,” Chapter I X , Contract AT-(40-1)-1313 with the University of P a r t 2 , “Reaction Kinetics of Enzyme-Sub3trate Compounds,” Arkansas. The assistance of Dr. G. T . Johnson in edited by A . Weissberger, Interscience Publishers, Inc., New York, handling cultures of Azotobacter is acknowledged. N. Y . , 1953. (15) M. D. Kamen in “Inorganic Ritrogen Metabolism,” edited by W. D. WcElroy and B. Glass, Johns Hopkins Press, Baltimore, hld., 1956. (16) A. Harrison and H. G. Thode, Trans. Faraday Soc., 64, 84 (1958). (17) W. A. Bonner and C.J. Collins, THISJ O U R N A L , 76, 4616 (1953)

FAYETTEVILLR, ARKANSAS (18) G . E. Hutchinson, “ T h e Solar System,” Vol. 11, edited by G. P. Kupier, Chapter 8, “ T h e Biogeochemistry of the Earth’s Atmosphere,” University of Chicago Press, Chicago, Ill., 19.54. (19) R. S.Scalan, P h . D . Thesis, University of Arkansas, 1958.